Pembrolizumab (pembro) monotherapy for patients (pts) with high-risk non–muscle-invasive bladder cancer (HR NMIBC) unresponsive to bacillus Calmette–Guérin (BCG): Results from cohort B of the phase 2 KEYNOTE-057 trial.

LBA442 Background: Although most pts with HR NMIBC respond to BCG, pts whose cancer does not respond or who relapse within 12 mo have poor prognosis and require radical cystectomy (RC). The single-arm, multicohort phase 2 KEYNOTE-057 trial (NCT02625961) was designed to investigate the safety and efficacy of pembro monotherapy for pts with BCG-unresponsive HR NMIBC (per FDA) who were ineligible or declined to undergo RC. Results from cohort A (carcinoma in situ [CIS] ± papillary tumors) showed a clinical complete response rate of 41% at 3 mo and led to approval of pembro monotherapy for such pts in the United States. We describe the results from cohort B (papillary tumors without CIS). Methods: Pts were aged ≥18 y with BCG-unresponsive HR NMIBC with papillary tumors only (high-grade Ta or any-grade T1) at baseline and ECOG PS 0-2. Pts received pembro 200 mg every 3 wk (Q3W) for ≤35 cycles (~2 y). Cancer was assessed at 12 wk and Q12W thereafter if no recurrent HR NMIBC or progression was observed; CT urography was done Q24W. Primary end points for cohort B were 12-mo disease-free survival (DFS) rate of HR NMIBC as assessed by central pathology/radiology review and safety, assuming a 12-mo DFS of >20% for HR NMIBC. Secondary efficacy end points were 12-mo DFS rate of any disease; progression-free survival (PFS) to worsening of grade, stage, or death; PFS to muscle invasion, metastasis, or death; and overall survival (OS). Results: Overall, 132 pts received pembro for a median of 9.5 cycles (range, 1.0-35.0). Median age was 72 y (range, 37-87); 57 pts (43.2%) had T1 stage; all pts (100%) had urothelial histology; 104 pts (78.8%) were male; pts received a median of 10 (range, 6-33) prior BCG instillations. Median follow-up was 45.4 mo (range, 14.9-77.1). Efficacy data are shown in Table. Thirty-one pts (23.5%) had RC after stopping pembro. Treatment-related AEs occurred in 97 pts (73.5%); 19 (14.4%) had a grade 3/4 treatment-related AE and 14 pts (10.6%) discontinued due to a treatment-related AE. No deaths from treatment-related AEs occurred. Conclusions: Pembro showed notable antitumor activity in pts with BCG-unresponsive non-CIS papillary HR NMIBC after ~45 mo of follow-up. Toxicity was manageable and consistent with that in cohort A, with no new safety signals. Results suggest pts with non-CIS papillary HR NMIBC unresponsive to BCG who declined or were ineligible to undergo RC may also benefit from pembro monotherapy. Clinical trial information: NCT02625961 . [Table: see text]